JURNAL TEKNIK INFORMATIKA DAN SISTEM INFORMASI
Vol 12 No 2 (2025): JATISI (Jurnal Teknik Informatika dan Sistem Informasi)

Prediksi Keberhasilan Adopsi Rapor Digital Madrasah Pendekatan Terpisah menggunakan Algoritma Naive Bayes dan Decision Tree dengan Pruning

Jamaluddin, Jamaluddin (Unknown)
Yonhendri, Yonhendri (Unknown)



Article Info

Publish Date
12 Jun 2025

Abstract

MTs Negeri 1 Batam as one of the Islamic educational institutions in Indonesia is a part that has adopted digitalization. Rapor Digital Madrasah is one form of digitalization launched by the Ministry of Religious Affairs as an effort to modernize the academic reporting system. This study aims to predict the success rate of Madrasah Digital Report Card adoption using different machine learning approaches, namely Naive Bayes algorithm and Decision Tree with pruning. The approaches were separated through each algorithm's specific preprocessing pipeline, with Naive Bayes using numerical feature standardization and Decision Tree without standardization. Results showed Decision Tree performed higher in each evaluation metric, with 98% vs. 92% accuracy and 100% vs. 83% minority class recall. Feature importance analysis identified three main factors influencing successful adoption, namely Technical Constraints (38.4%), Frequency of Use (24.7%), and Digital Experience (18.2%). Overall, the adoption of digital report cards was successful with a success rate of 86.3%. The results of this study can be implemented in the development of more effective and targeted intervention strategies in the future in increasing the success of Madrasah Digital Report Card adoption throughout Indonesia.

Copyrights © 2025






Journal Info

Abbrev

jatisi

Publisher

Subject

Computer Science & IT

Description

JATISI bekerja sama dengan IndoCEISS dalam pengelolaannya. IndoCEISS merupakan wadah bagi para ilmuwan, praktisi, pendidik, dan penggemar dalam bidang komputer, elektronika, dan instrumentasi yang menaruh minat untuk memajukan bidang tersebut di Indonesia. JATISI diterbitkan 2 kali dalam setahun ...